Accounting for surface reflectance anisotropy in satellite retrievals of tropospheric NO2

Atmospheric Measurement Techniques (Impact Factor: 3.21). 09/2010; 3(5):1185-1203. DOI: 10.5194/amt-3-1185-2010

ABSTRACT Surface reflectance is a key parameter in satellite trace gas retrievals in the UV/visible range and in particular for the retrieval of nitrogen dioxide (NO2) vertical tropospheric columns (VTCs). Current operational retrievals rely on coarse-resolution reflectance data and do not account for the generally anisotropic properties of surface reflectance. Here we present a NO2 VTC retrieval that uses MODIS bi-directional reflectance distribution function (BRDF) data at high temporal (8 days) and spatial (1 km × 1 km) resolution in combination with the LIDORT radiative transfer model to account for the dependence of surface reflectance on viewing and illumination geometry. The method was applied to two years of NO2 observations from the Ozone Monitoring Instrument (OMI) over Europe. Due to its wide swath, OMI is particularly sensitive to BRDF effects. Using representative BRDF parameters for various land surfaces, we found that in July (low solar zenith angles) and November (high solar zenith angles) and for typical viewing geometries of OMI, differences between MODIS black-sky albedos and surface bi-directional reflectances are of the order of 0–10% and 0–40%, respectively, depending on the position of the OMI pixel within the swath. In the retrieval, black-sky albedo was treated as a Lambertian (isotropic) reflectance, while for BRDF effects we used the kernel-based approach in the MODIS BRDF product. Air Mass Factors were computed using the LIDORT radiative transfer model based on these surface reflectance conditions. Differences in NO2 VTCs based on the Lambertian and BRDF approaches were found to be of the order of 0–3% in July and 0–20% in November with the extreme values found at large viewing angles. The much larger differences in November are mainly due to stronger BRDF effects at higher solar zenith angles. To a smaller extent, they are also caused by the typically more pronounced maximum of the NO2 a priori profiles in the boundary layer during the cold season, which make the retrieval more sensitive to radiation changes near the surface. However, BRDF impacts vary considerably across Europe due to differences in land surface type and increasing solar zenith angles at higher latitude. Finally, we compare BRDF-based NO2 VTCs with those retrieved using the GOME/TOMS Lambertian equivalent reflectance (LER) data set. The relative differences are mostly below 15% in July but in November the NO2 VTCs from TOMS/GOME are lower by 20–60%. Our results indicate that the specific choice of albedo data set is even more important than accounting for surface BRDF effects, and this again demonstrates the strong requirement for more accurate surface reflectance data sets.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The recent increase in spatial resolution for satellite instruments has made it feasible to study distributions of trace gas column densities on a regional scale. For this application a new gridding algorithm was developed to map mea-surements from the instrument's frame of reference (level 2) onto a longitude–latitude grid (level 3). The algorithm is designed for the Ozone Monitoring Instrument (OMI) and can easily be employed for similar instruments – for example, the upcoming TROPOspheric Monitoring Instrument (TROPOMI). Trace gas distributions are reconstructed by a continuous parabolic spline surface. The algorithm explicitly considers the spatially varying sensitivity of the sensor resulting from the instrument function. At the swath edge, the inverse problem of computing the spline coefficients is very sensitive to measurement errors and is regularised by a second-order difference matrix. Since this regularisation corresponds to the penalty term for smoothing splines, it similarly attenuates the effect of measurement noise over the entire swath width. Monte Carlo simulations are conducted to study the performance of the algorithm for different distributions of trace gas column densities. The optimal weight of the penalty term is found to be proportional to the measurement uncertainty and the width of the instrument function. A comparison with an established gridding algorithm shows improved performance for small to moderate measurement errors due to better parametrisation of the distribution. The resulting maps are smoother and extreme values are more accurately reconstructed. The performance improvement is further illustrated with high-resolution distributions obtained from a regional chemistry model. The new algorithm is applied to tropospheric NO2 column densities measured by OMI. Examples of regional NO2 maps are shown for densely populated areas in China, Europe and the United States of America. This work demonstrates that the newly developed gridding algorithm improves regional trace gas maps; its application could be very helpful for the study of satellite-derived trace gas distributions.
    Atmospheric Measurement Techniques 02/2014; 7:451-467. DOI:10.5194/amt-7-451-2014 · 3.21 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We present a new algorithm based on the iterative spectral fitting technique for direct retrieval of nitrogen dioxide (NO2) vertical columns from hyperspectral satellite measurements, and a new spatial technique for separating the stratospheric and tropospheric contributions to the total NO2 vertical columns. This direct vertical column fitting (DVCF) algorithm allows more complete treatment of underlying algorithm physics compared to the differential optical absorption spectroscopy (DOAS) technique, and enables more accurate accounting for the effects of spectral and altitude variations in NO2 measurement sensitivities. This DVCF algorithm and the stratosphere-troposphere separation scheme are applied to ultraviolet measurements from the Nadir Mapper of the Ozone Mapping Profiler Suite (OMPS) on the Suomi National Polar-orbiting Partnership spacecraft. The first results demonstrate the potential of OMPS for sensitive global monitoring of tropospheric NO2. Comparisons show good agreement with Aura Ozone Monitoring Instrument (OMI) tropospheric NO2 data, and that OMPS data have sufficient quality to continue and extend OMI NO2 data records.
    07/2014; 41(13). DOI:10.1002/2014GL060136
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: {textlessptextgreatertextless}br/textgreaterWe have conducted a feasibility study for the geostationary monitoring of the diurnal variation of tropospheric NO2 over Tokyo. Using NO2 fields from a chemical transport model, synthetic spectra were created by a radiative transfer model, SCIATRAN, for summer and winter cases. We then performed a Differential Optical Absorption Spectroscopy (DOAS) analysis to retrieve NO2 slant column densities (SCDs), and after converting SCDs into vertical column densities (VCDs), we estimated the precision of the retrieved VCDs. The simulation showed that signal-to-noise ratio (SNR) [greater-or-equal, slanted]500 is needed to detect the diurnal variation and that SNR [greater-or-equal, slanted]1000 is needed to observe the local minimum occurring in the early afternoon (LT13-14) in summer. In winter, the detection of the diurnal variation during LT08-15 needs SNR [greater-or-equal, slanted]500, and SNR [greater-or-equal, slanted]1000 is needed if early morning (LT07) and early evening (LT16) are included. The currently discussed sensor specification for the Japanese geostationary satellite project, GMAP-Asia, which has a horizontal resolution of 10km and a temporal resolution of 1hr, has demonstrated the performance of a precision of several percent, which is approximately corresponding to SNR=1000textasciitilde2000 during daytime and SNR [greater-or-equal, slanted]500 in the morning and evening. We also discuss possible biases caused by the temperature dependence of the absorption cross section utilized in the DOAS retrieval, and the effect of uncertainties of surface albedo and clouds on the estimation of precisions.textless/ptextgreater
    Advances in Space Research 11/2011; 48(9):1551-1564. DOI:10.1016/j.asr.2011.06.029 · 1.24 Impact Factor